Klasifikasi Penyakit Tanaman Jagung dengan Kecerdasan Buatan Berbasis CNN

Abstract

Quick and accurate detection of corn plant diseases is a challenge for farmers who require early treatment to increase crop yields. This research develops a corn plant disease classification system using Convolutional Neural Network (CNN) to differentiate between healthy and infected leaves. The dataset contains 341 images of corn leaves from Kaggle, divided into two classes: healthy and infected. CNN was chosen to recognize visual patterns in leaf images so that it is able to detect differences in conditions. The test results showed an accuracy of 97.10%, indicating excellent performance in corn disease classification. This CNN model is effective as an automatic detection tool, providing practical solutions for farmers in early detection of plant diseases, hopefully speeding up treatment, reducing losses and increasing agricultural productivity.